This paper compares intelligibility, ethnolinguistic identity, and linguistic structure between rGyalrongic lects in the following locations belonging to Ganzi Prefecture: Daofu County's Xianshui Town, and Danba County's Niega and Geshiza Townships. rGyalrongic language varieties, located in both Aba and Ganzi Prefectures of northwestern Sichuan, have been demonstrated to be a genetic subgroup within Sino-Tibetan by J. Sun's (2000) discovery of four paradigmatic, individual-identifying, and morphologically complex shared innovations. Although rGyalrongic has been established as a subgroup, the internal diachronic and synchronic divisions have not been rigorously established. J. Sun proposed three distinct synchronic languages belonging to rGyalrongic—rGyalrong, Lavrung, and Horpa-Shangzhai—but he also suggests that rGyalrong could be conceived of as three languages and Horpa-Shangzhai could be considered two languages. Jacques (Forthcoming) and Gates (2012) challenge the hypothesis that rGyalrong is a single synchronic language, the former proposing a four way division and the latter identifying five distinct synchronic languages. J. Sun's Horpa and Shangzhai have been granted ISO 639-3 codes [ero] and [jih] respectively, and were subsequently entered into the Ethnologue (Lewis 2009). This paper concerns itself with Horpa-Shangzhai, referred heretofore as the ‘Western rGyalrongic' language cluster. Western rGyalrongic language varieties, with about 45,000 speakers scattered throughout Ganzi Prefecture's Daofu, Luhou, Danba, and Xinlong Counties and Aba Prefecture's Rangtang and Jinchuan Counties, were first categorized as a single synchronic language known under the names Ergong (Sūn 1983) and Daofu (Huáng 1991), respectively. More recently, Suzuki (2009, 2010a, 2010b, 2012) has proposed that four synchronic languages should be recognized from this cluster, renaming Horpa as sTau (in Daofu, Luhuo, and Danba), Nyagrong Minyag (in Xinlong), and Geshitsa (in Danba County), and renaming Shangzhai as Puxi. Before any of the above competing hypotheses (or others unmentioned) can be chosen as the best reflection of the ethnolinguistic situation, adequate criteria must be agreed upon and sufficient evidence must be provided, integrating synchronic and diachronic, ethnocultural and linguistic. Fresh data used in this paper serves as a catalyst towards that end and simultaneously provides evidence supporting Suzuki's split of Geshitsa from sTau. In short, intelligibility testing was conducted by asking participants from Xianshui, Geshiza and Niega to listen to recorded sentences from each lect and retell what they understood. A detailed analysis of these retellings reveals low intelligibility between speakers of Xianshui and speakers of Geshiza/Niega, and marginal intelligibility between speakers of Geshiza and speakers of Niega. Ethnic identity was explored through the use of participatory methods and informal interviews, revealing ethnic affinity between speakers from the three locations, a mix of high and low perceived intelligibility, and onomastic identities that correlate with intelligibility testing results. In addition, wordlists, sentences, and natural texts from Xianshui, Geshiza, and Niega were collected and a preliminary analysis is presented, including a presentation of isogloss bundles that correlate with intelligibility and ethnic identity.